Sunday 30 January 2005

Comment to "Permanent Injustice: Rawls' Theory of Justice and the Digital Divide"

In the recently published Educational Technology & Society Journal (Vol 8 Issue 1), Elizabeth Hendrix argues that

Rawls’ theory of justice does not work in practice with regard to technology, or as a way to solve the digital divide and the inequalities in school funding. She argues that another ethical theory should guide technological funding and policies in schools, embracing theories by Levinas, Noddings, Davis, Freire, Nkrumah, and Buber, in order to open scholarly discussion on the issues of injustice and technological funding inequities.


I have no previous reading of any papers on "justice" and I declare total ignorance in this area. However, I do feel that whatever a funding policy may be designed, digital divide, or different levels of technology access by different population groups, is an unavoidable fact in life. Again, I am not arguing designing a "fairer" funding policy, I just want to remind ourselves that we should not take technology access as a given. For many, it is far from the true. See my previous post, The Fallacy of Digital Equality.

Online Jigsaw Class - an alternate use of Fablusi

I have posted a way of implementing the Jigsaw technique in my other blog: Asynchronous Collaborative Learning Activities.

Friday 28 January 2005

Educational Games Don't Have to Stink!

This is a wonderful article from Gamasutra. Free registration required to read the article.

After telling an experience during his graduate stage, some insightful comments appear:

A teacher's two greatest tools are charisma and attention, both things that computers cannot offer. A teacher uses his charisma to create interest and excitement in the student, and uses his attention to reward, punish, and compel attention back from the student.


My heretical view is simply this: computer games don't teach. I think the idea that you can teach using computer games is based on a flawed analogy between gameplay and learning. Here's how the analogy goes. Players of games have to overcome obstacles in order to achieve victory. They do this by learning the weaknesses, or limitations, of the opponents they face. Similarly, students learn knowledge in order to pass tests. So learning a fact is equivalent to defeating an enemy, and passing a test is equivalent to achieving victory. ...


In short, it's my belief that games don't teach, they illustrate. That's an important distinction. Games are not useless in the educational process, but they're not good at teaching per se. Games are good at creating understanding of knowledge the student already has. And they're excellent at transforming abstract ideas into concrete experience. Games don't teach, but they can help people learn.


With all these explained, here are some of the suggestions to create "educational games" from the author:
Admit that games don't teach, they illustrate.

Don't make games that are too much fun.

Don't make games that aren't fun enough.

Don't make games that take too long.

Don't make games that obscure the principles you want to illustrate.

Include advisors.

Don't forget the value of creative play.

Don't try to serve chocolate-covered broccoli.


These ideas are good ideas. Since my interest in more in collaborative type of learning (e.g. role play simulation), I cannot agree to one of the design guidelines here. For example, the duration of the game need not and should not be confirmed to the duration of a 50-minutes class. Let us bring the educational process into the learners' home. Let them learn at any convenient time. But, again, this is not the kind of game the author is talking about.

Thursday 27 January 2005

Asynchronous Collaborative Learning Activities

I have started a new blog which is focussed on " Asynchronous Collaborative Learning Activities". The first entry is related to how we may modify debate for asynchronous online use. Hope you enjoy this and please give me comment and support.

JSON and its implementation gotcha

Stephen points to a slashdot post about JSON (Javascript Object Notation) promoted by Douglas Crockford.

Fablusi replies heavily on JSON for its communication with the server and inter-module communication. However during our implementation, I discover a few issues which are not discussed anywhere else known to me. (Please point me to any discussion of JSON, if you know of any).

1. Internet Explorer generally fails to initialise an object in JSON notation. However, it loads an array of objects correctly. So if you want to initialise a JSON object in cross browser application, use this pattern:
thisObj=eval("[{'membersName':'MemberValue',...}]")[0];
i.e. transfer using an array instead of an object, or add the square brackets at the client side before conversion.

2. There are other even lighter weight mechanism for transferring a set of data of the same object. (by not repeating the membersName many times..., but require some software support at the receiving end.)

3. the evaluated object does not have any methods. It is data only. This can proved to be quite difficult to use the object effectively.

4. Eval is a depreciated Javascript method. :-(

Current Fablusi v2 users will notice that the save format of the simulation data is in traditional Javascript object initialisation format - NOT JSON because I need the initialised objects to have methods defined in the system.

While JSON is a good idea and I look forward to seeing how JSON-RPC may work, use with an understanding of its limitation. The benefit (light weight) outweighs its limitation.

Tuesday 25 January 2005

Meta Meta Meta Data Draft 0.2

Previous drafts are 0.1 and 0.1b.

For your viewing convenience, I have copied the essence of the 0.1 draft below:

Say we have a set of data, let call it S1 with elements s11, s12, s13, ... s1n. These elements was referred to as type 1 data in the previous paper.

Now, apply a "meta" operation* on each of the element in S1 which will produce a set S2 with elements s21, s22, s23, ... s2n where s21 is the metadata of s11. These elements ( s21, s22, s23, ... s2n) was referred to as type 2 data in the previous paper.

Note that elements of S2 are data as well. These elements are themselves type 1 data and hence we can apply "meta" operation on these as well to produce another set of type 2 data. This is infinitely recursive.

What is interesting, and perhaps confusing, is that there exist more than one meta operation. In fact, there are infinite number of meta operations. Each meta operation will produce a set of Type 2 carrying the implicit characteristics of the meta operation. We further define a meta-meta operation as an operation on a set S which will extract the common characteristics all elements in the set S2 to produce M1. Since there are infinite number of possible meta operations, there exists infinite number of characteristic, M1, M2, M3,... Mn,... Each of these characteristics, when expressed in as data, is what we refer to as type 3 data in the previous paper.

An example of Mi may be the Dublin Core specification, which defines a particular meta operation. The process of producing M1 is the meta-meta operation. Different community of practice will obviously have their own variations of meta operation (adoption and extension of DC) producing Mi.

Meta-meta operation applies on data elements. Since type 1 data is data, we can also meta-meta operation on type 1 data. One of the possible characteristics of type 1 data is the link information among the elements. This link information has been an important information to determine the "page-rank" in Google's search result. Again, there are other meta-meta operation which can be applied to type 1 as well as type 2 data.

The utility of this data model may be used to understand the work in metadata....

*A meta operation may be the extraction of metadata from the type 1 data, such as assigning dc.creator (in type 2 data of this post) to the value "Albert Ip" (for this particular post, which is type 1 data in this case). Another meta operation may be to create the frequency count of all the words used in this post.


David Gibson has sent a comment regarding the idea (on Mon, 24 Jan 2005 09:18:13 -0800 (PST)):

What prevents there being an infinite number of meta-4's (pardon the pun). You have defined up to the meta-meta-meta right? It seems that third level is mining the metadata schemas. And it seems that one could keep positing operations on the previous level. Why does it stop with 3 levels?


Thank you, David, for this very thought-provocative comment which helps me clarify some points in this concept.

The first meta operation (which applied on S1) is like the "traditional" metadata extraction. The operation applies on the individual elements of the Set S1. This model recognizes that there are many different types of meta operations, including the variation of the metadata schemes, or fundamentally different alogrithm, such as using inverted index.

We can say:

S2 is the result of operation m on S1 if and only if

for each element s2i in S2, s2i=m(s1i)


The meta-meta operation (or should we write it as Meta operation) operates on the set S1 to provide a datum m1 (was M1 in draft 0.1). Again, we note here that there are many different M operations resulting in m1, m2,...,mi,... which collectively are members of the set M1.

Since M1 is a set of data, we can apply meta-operation on it to produce a set of metadata for each element of M1. Alternately, we can apply a Meta-operation on the set M1 to produce one datum representing a common characteristic of the set M1.

I think the last paragraph here is the unclear point in draft 0.1.

Sunday 23 January 2005

The Fallacy of Digital Equality

Digital divide existed, exists and will continue to be there no matter what best effort we put in.

I am stating a fact, not advocating that we should create greater inequality. On the contrary, by recognising that digital divide is an inevitable fact, we should learn to develop learning for ALL, not only those best equipped.

Technology is advancing too quickly. Even with unlimited resources (in the unlikely political situation that the government is willing to fund all under-privileged areas to "catch-up" with the other areas in technology), the rolling out of the technology will take time. Some will get better technology *before* the other. That may be 1 year or two. This time different is already one or two generation difference in the technology. But the time the last region gets "updated", the other region has already fallen behind. If you look slightly further, in the un-developed world, they are not talking about education. They are struggling to get by the day without feeling thirsty or hungry!

I am not proposing to solve the world's problem. I just want to remind myself that inequality is a matter of hard life.

Developers are typically equipped with fast machine, huge amount of RAM, unlimited bandwidth, largest and highest resolution monitor. It is quite easy to get used to these environment. In the seeking of job satisfaction, it is easy to fall into the trap of using "cool" technique which requires in-proportionate amount of computing power just to display something irrelevant to the learning objectives. I fell into the same trap myself - just have to constantly remind myself that my philosophy is to give everyone opportunity to learn from your own work - not just those having the greatest technology.

But I am still an e-learning participant.

Wednesday 19 January 2005

10,000 visitors

This blog will welcome the 10,000th visitor today. Thank you for your support. If you are the lucky 10,000th visitor, please drop me a note. How to know if you are the 10,000th visitor, scroll down a little to look at the site counter, please.

I started this blog in August 2004. The past 6 months have been an incredible learning journey for me. I have learnt a lot, sharpened my ideas and made a lot of online friends. It is amazing to look back and see how much I have improved in my own thinking. This is a vivid illustration of the hand-on approach of self-learning and self-reflection.

I learnt my computing skill the same way. I entered the world of online role play also in a similar way. Learning and exploring as I go, making small discovery, progress and just continuous to amaze oneself, and be amazed, by new things that we discover. The writing of the blog helps to reflect on the ideas and helps to internalize the idea. What a great way for self-development, especially you know that you are sharing this with a lot of people.

Welcome to my 10,000th visitor. You are truly amazing!

10,000 visitors

This Blog will welcome its 10,000th visitors today. Thanks for all the support! If you are the lucky 10,000th visitor, please drop me a note.

Since I started blogging in August 2004, I have learnt a lot of things, sharpened my thinking in many areas and made many friends on line. This has been an incredible journey in such a short span of time. This illustrates how effective hand-on approach is to learning and self-reflection.

Tuesday 18 January 2005

Business Case for LMS and scalability design considerations

This post have two points to make:

This afternoon, I have lunch with two best friends from two different Australian universities. Both are middle managers and both are responsible for making a recommendation to the Senior Management for a campus-wide course delivery system. Eventually, they both will be responsible for the implementation and maintainence of the system. Both are non-technical and come from an academic background. After some wine and food, we started talking about LMS.

I was in favour of open source and have been pushing hard to both to adapt/adopt the MIT/Stanford open courseware and saki approach. However, both rejected the idea basing on their perceived value of their respective higher management. While one of the university has already made public their choice, the other's choice is still officially private information. But I know that they ended up choosing differently from WebCT and Blackboard. These two universities, due to their different background have chosen different commercial LMS. One of the strongest reasons for the choice is the perceived support offered by the vendors and the "additional" cooperation offered to them. They were both offered position in the development committee of the product so that their implementation experience and needs can be addressed in the product. Another interesting result of the negotiation was that they both got a 5-year license with a cap on the license fee.

It is true that LMS has become the tier-1 IT infrastructure that Australian Universities will rely upon. They equal the LMS to email system. They are happy to find a solution for the 80% of the academics and let that innovative 20% to find their own ways.

Content management system was considered together with the LMS. One of the University chose one because the other vendor configured their system wrongly during pilot. It seems that if configured correctly, the missing vendor can provide exactly the same feature. Because the CMS and LMS was offered for testing from two different servers, some of the functions they wanted failed. If the vendor has run the two systems on the same machine, it would have met all the requirement and remain a sound contender!

I saw a fundamental fault in the architecture if this is the case. I know where the problem is and I have a solution. But I am not disclosing this here. If any LMS/CMS vendor is reading my post, expect to pay good money for me to reveal this. :-) OK, some nice words and good food may do the trick too.

I like to add to the complex mix of factors scalability and fault-tolerant. Scalability can be obtained vertically and/or horizontally.

By a vertical scalability - I mean obtaining more throughput using more powerful hardware. Typically, if a system is not providing sufficent throughput, the institute replaces the hardware with a faster and bigger one. This is not a bad approach given the continuous Moore's law of increasing computing power per unit cost. But this is a wasteful approach. In many cases, the decommissioned hardware is used to serve second tier services, or not as demanding service.

Another scalability is horizontal scalability, or Google scalability. Instead of a single (or a few) powerful hardware, Google uses thousands off-the-shelf PC running in parallel. If more capacity is needed, Google adds more PC. (One of the problem facing Google is the tremendous amount of power required to run this army of PCs for the same computational throughput. I am not sure about this remark, but I inclined to believe that it is true!)

Inherent to this scalability issue is fault tolerant solutions. Redundancy is the best approach to provide fault tolerant to critical systems. Running multiple expensive big server in parallel just to provide fault tolerant is expensive. On the other hand, off-the-shelf PC are inherently error prone, but they are relatively easy and cheap to replace.

I am more in favour of Google's approach.

This implies a different approach to software architecture in building the application. Stating the obvious again, developing an application to run on many nodes and expecting any node to fail any time, but continuous to provide a non-stop service is quite different to developing the same application which based on a single never-to-fail computing node.

Ok, I cannot avoid coming back to Fablusi. For Fablusi, I have put into the software architecture the capability to run in parallel an army of PC as servers. First, the computational intensive part is shifted to the client side. The server is basically a huge storage and serve whatever required by the client. Second, each connection to the server is "stateless" meaning that any responding server would have sufficient information from the client request to execute the needed operation. My hope for Fablusi's future is a single unified platform to serve ONE single pedagogy to anyone connected to the Internet - the "Google" of role play. I understand that role play simulation is just ONE of many powerful pedagogies. I just hope that it serves the purpose we built it and serves it well.

The problem I mentioned earlier about the integration between LMS and CMS is very real and faces Fablusi too. If Fablusi is to integrate with other pedagogical approaches in a course, it must be able to function across server boundaries and domain boundaries. The solution has already been developed and in the public domain. It is just a matter of identifying the problem and applying the solution.

Simple Answering Machine for Skype Users

via incorporated subversion (thanks James)

My skype has now installed an answering machine. So if I am not with my computer, you can still leave me a message on my skype. That's great!

I am not advertising my skype name here. But if you have something to discuss with me, feel free to send me an email and I'll get back to you.

Monday 17 January 2005

Soundseeing to Sightseeing: Good or Not?

AMY GAHRAN of Contentious is doing an experiment:

She asked:

... about soundseeing expeditions – which are audio recordings made in the field which capture the ambient sounds and the creator’s descriptions. Soundseeing sets a context intended to create a vivid picture in the listener’s mind.

How crucial is that mental picture to the soundseeing experience?


This echoes well with my earlier post Learning Context: Do we need to render it?, our mind fills in the missing details from our past experience.

So I went ahead and listened to the soundseeing. This is my imagined scene (briefly):
I heard the wood cracking sound from the wood fire. I imaged a room in reddish colour, Amy lying cosily in some soft fabric in front of the wood fire and typing away. I did not pay much attention to her description of the car junk yard - there is no ambient sound to support that!

My reaction to Amy's invitation is that I would rather have not seen the photographs. With the exception of the one where Amy was typing away on her bed, most of the photos do not depict the image I have formed from the soundsighting. When rendered, it totally destroyed my mental image.

Again, rendering the learning context is not a matter of right or wrong. It is a matter to do with the purpose. If this is a designed experience for learners and if we like to promote imagination and get the learners to focus on the important issues (except when the issue is visual!), rendered environment may not help.

Fangs: The Screen Reader Emulator Plugin for Mozilla

via EdTechPost

so many people simply assume that if it conforms to W3C WCAG or Section 508 guidelines, that's enough. It's not, and it's a case where 'seeing' (actually 'hearing') is believing. 'Fangs' allows you to read web pages more like how the users of an assistive reader will hear them.


Totally agree. Accessbility is much much more than just conforming to a standard. A tool like Fangs will help developers who are not visually impair to understand the problems of their screen layout. I look forward to installing this plug-in and I think it is a good idea to promote it.

Saturday 15 January 2005

Learning Context: Do we need to render it?

aka: Online Role Play Simulation Design - 4
Previous posts in this series: 1, 2 and 3.

Working almost against the main trend, Fablusi, the role play simulation platform I develop is NOT "pedagogically neutral". In fact, it embraces one pedagogical approach, namely problem-based, experiential, online role play. Roni and I called it "dynamic goal base" - meaning the game goal within the role play is changing all the time. We spend a lot of time thinking how our platform can deliver the kind of learning we designed into the platform and how effective that may be. One issue which came up since the LOW-1 meeting in Finland last year is how "virtual reality" rendered world can be integrated into Fablusi. Before answering that question, I am asking: "Does rendered world enhance the kind of learning that we build into Fablusi?" Here is my answer at this point in time.

Here are several assumptions/beliefs/facts I have about human nature:
1. Human sensory system is highly selective. Depending on the focus at the time when the information arrives, only those information which have our attention are captured. Most, if not all, other information are simply got filtered out. (This has been verified by a famous HCI experiment - but I cannot remember the source.)

2. Our internal world model is built, bit by bit and accumulated throughout our life via a lossy memory system. Such accumulated experience has been filtered by our sensory system. (This is very much the basis of constructism.)

3. When a model is lacking in details, our brain will fill in the details when necessary. This is what I called "imaginated reality". I can recall how disappointed I was when one of the character of a novel was played by one of the most beautiful actress in a television show. Although the actress is beautiful, she cannot compared with the imagined beauty I have created via reading the novel. One interesting thing about the "details" we filled in. Many may be based by experiences which have escaped our initial "attention" but somehow retained and show up. Others are simply made up at the "blink".

The way learning context is created by Fablusi online role play:

Our online role play mainly deal with human situations which involve different roles playing different stakeholders under different social relationship. The context of the "game" is abstracted to a number of roles representing different interest of different stakeholders. The players get a brief role information describing the situation of the game from the role's point of view. The game is started by the release of the initial kick-start episode. These roles interact in iSpace (interaction space). By differentially giving the roles different level of information, different rights in various iSpaces and wealth, we created a framework to model most social situation.

The role information is deliberately brief. One of the reason is to enable the players to embellish the role, filling in some details about the roles. Obviously, no one will be able to describe a full persona in the limited number of words allowed (by time and by design). However, this created a springboard for further imagination and players can fill in details from their past experience - and hence link the present simulated experience to their existing world model.

For political science simulations, where stakeholders are real political figures, we do not need to create the images of the persona in the game. The players can map their roles to real world figures.

For non-political simulations, e.g. commercial or human resource simulations, the players usually create a persona typical to their past experience. The imagined reality among the players are obviously different, but just like any other human interaction, we never have a perfect model of our communication partner any way. In this way, we have created a very realistic simulation, linking players past experience right into the game - AND that's exactly what we want to achieve.

iSpaces, e.g. corporate meeting rooms, the News room, the UN council are abstracted to asynchronous interaction spaces. We do not provide extensive graphics to represent the iSpaces as well. The players "tele-port" from one iSpace to another simply by clicking the appropriate buttons. There is no "walking" between iSpaces.

One obvious suggestion was to use 3D-rendered world to represent the iSpaces. I am rejecting this based on three reasons:
1. Typical interaction in 3D-world is synchronous. This will break some of the pedagogical values we have implemented using asynchronous mode, not to mention the additional loss in flexibility to the players in playing the simulation at any convenient time.
2. The extra rendering details may shift the attention of the players from thinking strategically and tactically how to deal with the simulated situation to the graphical details, which does not add any additional value to the learning objectives.
3. 3-D world rendering is an additional cost with no obviously benefit.

Another suggestion is to use 3D world to represent the "desktop" of the players. The reasons 2 and 3 from above applies.

Avatar may be used to represent the other roles in the game. Since we are leveraging on the asynchronous nature of our model to provide players "breathing space" to do background research, discuss with team members and formulate moves. Taking that away is not constructive to the pedagogical power of the design. Again, when you have an imaged model, any rendering may destroy your imaged model - which may not be in-line with our design as well.

Please note that my exposition here only relate to the value of virtual world in online role play. Again, my view of life is "fit for the purpose". Virtually rendered world must be good for something. Here I am just focussing on online role play simulation.

Tuesday 11 January 2005

Asynchronous Meetings Standing Order

In the course of our work on online role play simulation, we continue to discover a lot of differences between physical world and online world, especially when the online world is asynchronous. The following is NOT a comprehensive list of the differences, but significant enough to linger in my mind all the times related to online meeting especially related to online role playing.

Many meetings, especially those formal ones, will have rules of the meeting, or standing order, e.g. just a quick search on Google, I found this and this. Typically, such standing order will specify the time and method of calling a meeting. The timing for calling of election seems to be a powerful political tool to give your own party is highly winning chances (as in British and Australian Parliaments). One of the purpose of this fixed schedule of calling a meeting is to give participants sufficient notice in order to present in the meeting. Obviously, in an asynchronous meeting, the meeting time is an extended period and such a requirement becomes unnecessary. Actually, I have seen some Fablusi players spent significant amount of time and effort to "arrange a meeting" where in fact our analogy is a continuous meeting in the designated iSpace (interaction space). Of course, if they want to use the chat feature, a common acceptable time has to be arranged before hand.

Another common feature in Standing order is the requirement of a "Quorum" before the meeting may commence. Obviously it is to ensure that decision and discussion is reached and carried out within a "significant" representation of the body in which the ruling of the meeting may apply. Again, this requirement is needed in a physical meeting and can be relaxed in an online asynchronous meeting.

Business in the meetings are discussed and resolution reached in a pre-defined order - the agenda. Any change of the agenda also requires additional formal motion and voting. This is an aspect we try to simulate in Fablusi so that some roles may make use of the agenda to arrive at some desired outcome, if such roles have such right.

Limited by time, most meeting will also have limitations of time allowed for each participants to "have the floor". Again, this is a requirement that we can relax in an online asynchronous meeting.

One of the greatest difference is the way voting is conducted. In physical meeting, because of the attention span, when there is a motion on the table, the discussion will be focussed on the motion. Any other discussion is not allowed. There is only one motion on table. If there is any amendment, the motion is typically suspended and the discussion is focussed on the amendment itself. Again, this is carried out one at time.

I don't see that we should apply the same restriction on the online meeting environment. Hence, in Fablusi iSpaces when voting is allowed, simulation author can allow multiple motions can be put on table with multiple amendments at the same time. By using threading, discussion on each of the motion and amendment can be carried out in parallel. We are yet to see how this may influence the quality of online meeting. I am certainly interested and will keep a keen eye on such meeting.

Just like role playing, when we implemented it online, we see different limitations and potentials. I believe there is still a long way before we can fully understand how to conduct online meeting effectively. Comments welcome.

Monday 10 January 2005

My wish for the future of e-learning - with a businessman hat on

Is content king in e-learning industry? May be a doubtful YES today, it is a definite NO in a slightly longer term.

The marginal cost of reproduction of digital content is near zero. The upfront cost of course production may be huge and there may be copyright protection put in place by the copyright owners. This will not stop other people doing the same. Note that copyright only protect the manifestation of an idea. It does not prevent anyone else to create another manifestation of similar or same idea (this is the difference between patent and copyright, right?) As authoring tools advance, the cost of production will continue to decrease. The value of digital content will quickly reduce to near zero by the nature of market competition. I was told that the cost of "English as a second language" course in China have dropped 70% per year for 5 years at an AusWeb conference years ago. That's the result when people realise that ESL in China has a market and suddenly everyone jumped in. There are other initiations which drive the cost of content down quickly, .e.g MIT OpenCourseWare, BBC open-source streaming of its video content, the highly successful wikipedia, Google's digital library, project Gutenberg, etc.

For the "birthday cake" maker, see my birthday cake analogy, my advise is:

If you are a good content provider, you can still capture a good part of the value-chain.


Unlike physical goods whereas when a product has established a market, people will repeatedly buy the same product again without any need of "discovery", digital content is typically a use-once product (exceptions in this include songs, music and other "artistic performances", but informational content is very much a use-once product.) Another viable business in elearning is content discovery services. However, I am not sure a direct user pay model is a viable business model. May be Google's advertising-based model. This is the filling station part of the e-learning business.

On the other hand, organisations are waking up to realise that "Return of Investment" is enhanced when there is "real" change in attitude, behaviour and performance. Delivering training is NOT the same as delivering of information. Just-in-time bite size information is useful in the job-support situation. Again, the "filling station model" may be best approach to meeting this need.

My ultimate interest is still looking at training and education as a service industry where we provide valuable, memorable and life-changing experience to the learners. First person and second person experience are accidental (in the sense that it is unlikely to be planned) and most likely from informal learning channels. By encouraging active participations in blog type of sharing, community of practice and experience exchanges, the industry can provide a significant informal life-long education services. I will leave it to my readers to figure out a sustainable business model for informal learning. For planned (or formal) learning, a pure third person experience approach, while the cheapest (???) and easiest (???) to develop, may not be the best approach given the changes of this and coming digital generation. This is why I am betting everything I have on promoting simulated first person experience.

I don't gamble, but when I do, I bet my life on it.

Saturday 8 January 2005

Folksonomy and Taxonomy

I have been so busy lately that this debate has remained out of my radar. There is a debate currently on the merits of folksonomy and professionally developed taxonomy (or controlled vocabularies) for assigning values to metadata (tagging in the folksonomy parlance). The discussion focuses mainly on the information classification, which indeed is an important part of metadata based search. Many to Many has a post on the positive side which referred to a negative side. I believe there must be more posts on this debate.

My take on the issue is that we should not forget that metadata is only one of the many methods that we can use to find information, including e.g. digital images. See my earlier draft: Meta Meta Meta Data Draft 0.1b - in plain English For example, I do not believe Google has employed an army to tag images found on the web. It is likely to tag the images by the words near the image, or the file name of the image.

My pragmatic approach is still based on "fitness for the purpose". In a digital library where you have an army to classify information against a well established taxonomy. Why forces to use the evolving folksonomy? On the other hand, in evolving domains where developments are rapid and classification is fluid, taxonomy will not have been developed. Other methods should be used. Among the methods, folksonomy is one of them.

Discovery alone is not the best approach either. Without ranking, over thousands of returned results are quite useless. Even with progressive search refinement, still any use of the search result will demand quite an effort to work through a non-ranked list. Ranking of search results is NOT related to classification. Many methods are available too.

I would rather like to broaden the folksonomy verse controlled vocabulary debate to identify typical use pattern and debate on the best approach to meet the needs of "most" uses - leaving the long tail for niche application to fill the demands. Of course, one may choose to be a niche solution provider.

Rapid E-Learning: A Growing Trend

via Learning Circuits

I agree with the author that there are at least two expectation changes:
1. rapid development of e-learning (whatever what that means), and
2. lower cost of development for effective learning/training.

Here I would like to draw the attention to the different needs under different situations:

In a work support environment, we like to have just-in-time, easy to find information on how to accomplish the task at hand. Typically, the right bite of information at the right time is the most appropriate. The information consumer needs no more and no less information - just enough to solve the problem at hand efficiently and wants to move on to the next task. I suppose such kind of bite size training and support needs is better served by an efficient database with good search technology for locating the solution.

When there is a product launch which involves some training of a distributed sale force, I believe it would involve in-house rapid content development. The technique mentioned in the Learning Circuit article would meet most of the requirement.

When we are involved in changing the organisation culture to meet new business environment, this is a totally different kind of training and attitude modification requirement. A information shovelling model would not work. One cannot ensure that
1. the people in the organisation would even read the information,
2. if they really read it, whether it can actually change any attitude is problematic.

Yes, attitude and belief modification cannot be accomplished in a day or two. However, in this hyper-competitive business environment, the faster you can effect the change, the better is the opportunity of survival. Role play simulation, as I have argued earlier, is an effective tool for such program. The issues of using role play are time in the development of an effective RPS, and the cost of running the RPS.

Fablusi is text-based. The development cost of Fablusi-based training is small. The average cost of e-learning solution, according to the article, is US10,000 to 50,000 per hour. In comparison, Fablusi's cost would be next to nothing. For around US2,000 to US10,000, we can develop a RPS catering client's needs which runs for around 3 weeks and delivers life-changing experiences to the trainees. (Ok, there is an on-going fee based on per participant if you run the RPS on our server, but again it is very small compared to the traditional e-learning program. Getting a Fablusi server to serve your own need would cost a few hours of traditional e-learning development cost - but provide you with a continuous capability to run your attitude changing programs throughout your organisation perpetually.)

Another good news is that we are going to develop "Fablusi Lite", a light version of the Fablusi server which aims to enable a 5-minute development of RPS. Fablusi version 2 has taken us over 2 years to develop and has slipped the release date by about 6 months. I have under-estimated the amount of development work and over-estimated my ability! However, based on Fablusi version 2 core, Fablusi Lite is planned for release at the middle of this year. If you want to prepare yourself to enter this exciting training environment, please download the design worksheet and have a go with it. I am sure your time will not be wasted.

Friday 7 January 2005

Role and stakeholder in Role play simulation

In the version 3 role play simulation (RPS) design worksheet, Roni pointed out the distinction between roles and stakeholders. At other occasions, people asked,

In designing RPS, should I use a generic title (such as managing director) or give the role a name, e.g. Mr. Wealth Goldstein?


Any answer to this question has to meet the design objective? What are the purposes of the RPS and who are the target players (or learners) of the RPS?

The world is never a single dimension. Even a really elementary class of business school, students should be aware that a CEO is a human. Beyond the work place, s/he, like everyone else, is faced with other ups and downs in life - marriage problems, family commitments, health and so on. CEO also relaxes and goes to pub with colleagues. CEO also builds friendship, beyond the professional encounters at the workplace. By giving a role a proper name, interactions among persona in the RPS are more realistic and potentially have more learning opportunities to be discovered and leveraged upon.

This is this distinction to lead us to develop the independent concepts of "stakeholders" and "persona". In our terminology, stakeholders are viewpoints from a single professional related point of view. CEO, secretary, president, activists in Green movement and factory works are stakeholder view points. These present a set of public and private agenda, with influences from the circumstances in the scenario. Roles, as in a story, are characters in the RPS.

In order to create an effective role play, these stakeholder viewpoints are represented by personals in the role play (or simply roles). Roles have additional characteristics beyond representing the stakeholder viewpoints. Roles live in the RPS. Roles have personality. Roles has "fresh and blood".

Another important consequence of introducing this separation is that when we create a RPS, we concentrated on articulating the stakeholders viewpoint at an earlier stage, select those most relevant to learning objectives and create the scenario around typical issues. If we stop here, the RPS will be fairly "blend". By combining several stakeholders into a role, or by using several roles to represent a stakeholder viewpoint, the RPS sudden becomes much more playable. Players may have to fight the conflicting issues represented by the multiple stakeholders represented by the role, or need to align or fight for "stage time" between other roles who have similar stakeholder viewpoints. As the players bring in their previous personal experiences, the role play will develop differently at each run. Moderating/assessing a role play is not, and have never been, the same as the tedious process of reading the almost the same poorly written exposition of the same viewpoints as many times as the number of students in your class. AND the players learn better in authentic situation too.

One additional benefit of this design is the potential of creating roles with different levels of "importance": must have, critical, good to have and fillers. When combined with a flexible team size structure, this multiple levels of importance of roles created a large solution space for allocation of players into roles. The software, hence, can support the tedious task of allocating players into roles. This reduces a significant work load at the beginning phase on the RPS and will let teachers/instructors concentrate on their real job of leading the learners through the experience.

Wednesday 5 January 2005

Role Play Simulation design worksheet v3

The version 2.1 Role Play Simulation design worksheet, which has been online since May 8, 2003, has been updated by Roni Linser and is now available free. You need to supply a valid email address for the system to send the worksheet to you.

Although the worksheet aims to help authors to begin the creation of a role play simulation for eventual delivery using Fablusi v2 platform, previous versions have proved to be a useful tool for designing more general role play simulation.

The worksheet promotes a two step iterative design process. The first step, called the foundation phrase, guides you to think about the problem or set of problems/issues, places where stakeholders communicate with one another in confronting the issues, and a prioritised list of stakeholders involved in these issues.

You are then guided to fresh out more details including the key ingredients in creating "kick start episodes", the combination or splitting of stakeholders into roles, using interaction spaces to create a power relationships among the roles, and suggestions about creating tasks for roles.

Further customisation includes the issues with simulation time, look and feel, role playing with concept of wealth, role play for "knowing who" and so on.

Please feel free to distribute the design work sheet to whoever may benefit from this work or send them the URL (http://www.fablusi.com/requestDesignDoc.asp) for download.

Tuesday 4 January 2005

How Did Animals Escape Tsunami?

Wild animals seem to have escaped the Indian Ocean tsunami, adding weight to notions they possess a sixth sense for disasters, experts said Thursday.

Sri Lankan wildlife officials have said the giant waves that killed over 24,000 people along the Indian Ocean island's coast seemingly missed wild beasts, with no dead animals found.

"No elephants are dead, not even a dead hare or rabbit," said H.D. Ratnayake, deputy director of Sri Lanka's Wildlife Department.


If true, this is really amazing. Most news reports are, understandably, focussed on the human lost. Are there any further reports to support this?

Hard choices: researcher vs. blogger?

This post strikes a resounding resonance that I also struggled many years ago:

One of the comments I received from Inna Kouper on weblog conversation paper was about my involvement in it, the fact that as a researcher I could influence the conversation if you knew you were going to study it. I realised that Inna was right and included a statement that when the conversation was unfolding I wasn't going to study it.


So Lilia wanted to separate out the role of researcher and the role of a participant in a dialogue (blogger). Then...

...two weeks back I was struggling with another choice. I saw an interesting conversation unfolding, I wanted to participate, but I also thought that it would be a great "another case" to add to our paper since we discussed some future work with Aldo. Those two seemed to contradict: as a blogger I wanted to participate, as a researcher I knew that a better choice would be to stay away, so I could claim more objectivity in a future analysis.


Whoops! what is objectivity in the analysis of experimental studies? Later on, Sebastian Fiedler commented:

It is dead easy to cry for objectivity, reliability, etc. bla bla without ever exploring what this could and should mean in social research.

Of course, there are always proponents of one unified method for all sciences or one universal set of criteria of quality.

I have never understood why studying rock formations should and could follow the same set of rules like studying historically contingent affairs of human beings that are capable of self-reflection, change, etc.


I agree. As a Physics students over a quarter of a century ago, I learnt of the principle of uncertainty. The observation itself will distribute the state of the particle you are trying to observe. So "hard science" also recognised the dilemma...

...her being a long time participant in this particular social practice is putting Lilia in the privileged position to select especially illustrative materials to exemplify what she is trying to express.


The selection of episodes to include (or not to include) is a "subjective" decision (as oppose to objectivity)!

The question is of course, should social studies be criticised based on the notion of the researcher's objectivity or subjectivity? Unlike physical science, "social experiments" are difficult if not impossible (ethically as least). Social experiments are not repeatable. Social studies are NOT physical science and I don't believe the same set of rules (if any) should apply to judge the validity of reports on social phenomena .

Satellite Images of Tsunami Affected Areas

Aceh, Sumatra, Indonesia

Before (08 Dec 2004)


After (29 Dec 2004)


Large area of land just been wiped off...

This is heart-breaking...

Monday 3 January 2005

The only way is up

via Curiouser and curiouser. I did the clean sweep assessment p:6 + W:8 + M:9 + R:4 = 27.

I performed poorly in terms of relationship. The current heavy work load is hurting my relationship with my dearest. I hope I can repair this before it is too late...

On the positive note, the only way is up...

A sad end of 2004

It is sad to note that 2004 ended with such a huge tragedy - over 120 thousand people died in the Tsunami which occurred on 25th December 2004. Here is a blog which aggregates news feeds and blogs from a variety of sources around the world.

http://www.tsunami-info.org